Reply to “Murine Roseolovirus, Historically Known as Murine Thymic Lymphotropic Virus”

REPLY

In response to the comment on our recent paper (1) by Wang et al., we agree that there is a close relationship between murine roseolovirus (MRV) and a previously described virus known by several names, including murine thymic virus (MTV), murine thymic lymphotrophic virus (MTLV), and thymic agent, among others (2–4). As pointed out in our paper (1) and by Wang et al., the properties of MRV and MTV are very similar and include relatively unique phenotypes, including neonatal thymic necrosis, first described in 1961 (3). This topic was extensively discussed in our paper and mentioned in the abstract. Indeed, the title of our first submission of our manuscript to the Journal of Virology was “Mouse Thymic Virus (MTV) is a Murine Betaherpesvirus Closely Related to Human Roseoloviruses.”

While our interpretation was that MTV and the virus we sequenced, now termed MRV, are close enough to be considered to be one and the same, a reviewer of our original manuscript argued that unless we could definitively establish a relationship between the virus we sequenced and the initially published MTV, we could not make definitive conclusions about the relationship of our virus to MTV. To this end, we vigorously attempted to establish a relationship by several approaches during the revision of our manuscript. (i) Our viral stocks were obtained from a veterinarian who has since left academia. We contacted him to determine if the viral stocks he provided were derived from stocks that could be traced back to the original description of MTV or obtained from wild animals, but he no longer possessed these records. Since we did not have direct passage history from the original stocks, we could not know with certainty that our viral stocks were those of MTV. (ii) We contacted multiple authors of the last papers published on MTV (2, 5–9) but this pursuit was not fruitful, apparently because enough time had elapsed since these publications that investigators had apparently retired or were no longer actively working on MTV. Thus, no MTV stocks were available to us by this route. (iii) We contacted multiple individuals at Charles River Laboratories (CRL) because we became aware that CRL had developed diagnostic tests for MTV. We had hoped to obtain an aliquot of their MTV stocks, which were presumably used to validate their diagnostic tests, for sequencing studies. However, they required a material transfer agreement (MTA) that precluded publication of data from our intended sequencing of their stocks. When we emailed our CRL contacts with a request to amend their MTA, but we did not receive a response. Notably, we were communicating with CRL employees who did not place us in contact with Wang et al., and we thus had no knowledge of CRL personnel who were directly working with MTV or that CRL had a sequence of MTV. (iv) The literature mentions that the genomic sequence of the DNA polymerase of MTV is closely related to that of other betaherpesviruses (10), apparently allowing the International Committee on Taxonomy of Viruses (ICTV) to preliminarily classify MTV as murid herpesvirus 3 (MuHV-3). However, the sequence was not available in public databases, so we were not able to compare our sequences. Further investigation uncovered that the sequence mentioned by Ehlers et al. was obtained from the same veterinarian who contributed our MRV stocks, and thus, it also did not have documented direct passage history from the originally described MTV. (v) We contacted the ICTV for advice on whether our viral sequences would allow determination that MRV is indeed MuHV-3. However, because MuHV-3 had not been formally classified as a species by the ICTV, the organization could not provide judgment on this issue. Thus, we disagree with the premise of Wang et al. that we did not pursue a reference stock of MTV to determine the relationship of MRV to MTV.

Furthermore, as pointed out by the reviewer, the human roseoloviruses now include the closely related species human herpesvirus 6A (HHV-6A), HHV-6B, and HHV-7. At the time we submitted our manuscript, it was possible that MRV was related to MTV in a similar way; i.e., they were closely related but not identical roseoloviruses. In the absence of establishing a direct relationship between the virus stock we sequenced and MTV, we were not able to address the reviewer's concern and took the reviewer's advice to name the virus we sequenced, as mentioned in our paper (1).

The statement by Wang et al. claiming 100% sequence identity of their stock of MTV to our MRV sequence suggests that these viruses are actually identical. However, this statement is based on data not available for external evaluation. This claim is also surprising, as we did not sequence a cloned virus and found variation within our own published MRV genome, as noted in our paper. We encourage Wang et al. to either publish or deposit their sequence information in public databases and detail how the sequence was determined for evaluation by others.

Finally, we should point out that although most of the phenotypic characteristics of MRV infection are identical to those of MTV infection, we have some findings (S. J. Patel and W. M. Yokoyama, unpublished data) that diverge from the published literature. Although there are various explanations for these findings, one possibility is that the MRV that we now study has undergone genetic drift from the MTV studied in prior investigations. Thus, further evaluation is necessary to explore these issues, and a genomic sequence of MRV, such as that which we have reported (1) and deposited in a public database, will be helpful.